Hybrid Neural Network Model to Predict Stock Market Index: Evidence for the Tunindex Stock Market
Hybrid Neural Network Model to Predict Stock Market Index: Evidence for the Tunindex Stock Market
Author(s): Nadhem Selmi, Rabia AchairiSubject(s): Economy, National Economy, Financial Markets
Published by: Facultatea de Management, Academia de Studii Economice din Bucuresti
Keywords: Efficiency; Neural netwotk; Stock market index;
Summary/Abstract: In this paper we utilized as input variables the historical data and analyze the neural networks structure for financial time series forecasting. After chosen the better structure we compared the performance of the neural networks method and generalized autoregressive conditional heteroscedasticity GARCH model. In this paper we proved a combination of neural networks and statistical model is presented for forecasting the return TUNINDEX. This study motivated from the hybrid models (ARCH(2)-M-ANN, ARMA-ANN, GARCH-ANN) is better them the statistical process (ARCH, GARCH,…).
Journal: Business Excellence and Management
- Issue Year: 5/2015
- Issue No: 1
- Page Range: 79-88
- Page Count: 10
- Language: English